Survey (empiricism)

from Wikipedia, the free encyclopedia

As survey refers to the collection and analysis of data , for example in the field of expression , election or market research . In the case of a survey, the data does not have to be generated first, as in an experiment , in which an experimental situation must first be established before the data can be measured.

The (data) collection is intended to represent the characteristics of an object to be examined statistically. (Example: Survey of the voting behavior of a small group of persons entitled to vote in order to determine the position of the parties before the election (election survey)). Surveys can be made in writing, orally (also by telephone) or through observation.

A distinction is made between three types of survey:

  • the primary statistical survey (also primary statistics, field research ), the new just for this purpose of the survey data collected
  • the secondary statistical survey (also secondary statistics secondary survey uses), which only already pre-existing data for a new analysis target
  • the tertiary statistical survey, if only compressed data, for example in the form of mean values , is available.

In addition, a distinction is made between the basic survey (including full survey ) and partial survey . This involves examining a population that represents the entire population to be questioned (in the case of electoral surveys, for example, all eligible voters in a certain specified age group, or, for example, only the voting behavior of women, etc.). A sample is taken from the population, from which only part of the entire group is interviewed. The aim is to determine the behavior of the whole group according to certain statistical criteria. As a rule, samples are cheaper, less complex and less prone to errors.

Data collections also differ in terms of their bandwidth. The bandwidth of a data collection method is the amount of data that the method delivers as an end result. Broadband instruments (e.g. free interview ) are suitable for generating hypotheses, while narrowband instruments (e.g. concentration test) are more useful for testing hypotheses.

The use of empirical material is recommended in scientific work. With the help of this material it is possible to check whether the findings are up-to-date or to determine whether it is necessary to carry out new studies and to compare their results with those from the "old" empirical studies in the sense of a critical comparison.

Individual evidence

  1. ^ A b Ludwig Fahrmeir, Iris Pigeot, Rita artist, Gerhard Tutz: Statistics. The way to data analysis . 6th edition. Springer, Berlin 2007, ISBN 978-3-540-69713-8 , pp. 23–24 ( limited preview in Google Book search).
  2. ↑ Günther Bourier: Descriptive statistics. Practice-oriented introduction - with tasks and solutions . 10th edition. Gabler Verlag, Wiesbaden 2012, ISBN 978-3-8349-3270-9 , p. 27–28 ( limited preview in Google Book search).
  3. ↑ http://wirtschaftslexikon.gabler.de/Definition/erammlung.html
  4. ↑ http://www.sdi-research.at/lexikon/erammlung.html
  5. ^ Albert, Ruth: Empirie in Linguistik und Sprachlehrforschung, Tübingen, 2002, Chapter 3

Web links